Machine Learning Ph.D. Seminar

2009-2010 Schedule

September 15th, 2009 - Jason Weston, Google
How can we learn language from scratch?
Abstract
September 22nd, 2009 - Mehryar Mohri, Courant Institute and Google
Learning with imperfect data
Abstract
NOTE: This talk will be at Google, please contact Ameet by noon on Monday the 21st to register as a vistor.
September 29th, 2009 - Vasilis Gkatzelis, Courant Institute
Prediction, learning, and games, Chapter 1
This week we have a reading group meeting, attendees should have read the chapter and come prepared to discuss.
NYU students have free electronic access to the book here.
October 6th, 2009 - Ameet Talwalkar, Courant Institute
Large Scale Learning and Low-Rank Approximation: A Tutorial
Abstract
October 13th, 2009 - Slav Petrov, Google
Learning Latent Variable Grammars for Natural Language Parsing
Abstract
NOTE: This talk will be at Google. If you have not done so for a previous talk, please contact Afshin by Monday the 12th to register as a visitor.
October 19th, 2009 - Afshin Rostamizadeh, Courant Institute
Feature Selection for Learning
Abstract
October 27th, 2009 - Christos Papadimitriou, UC Berkeley
The Algorithmic Lens: How the Computational Perspective is Transforming the Sciences
No learning seminar this week, instead we invite you to attend a special talk in Room 102, WWH at 2:00pm
Abstract
November 6th, 2009 - 4th Annual Machine Learning Symposium
No seminar this week, instead we invite you to attend the New York Academy of Sciences 4th Annual Machine Learning Symposium on Friday November the 6th.
December 2nd, 2009 - Prediction, learning, and games, Chapter 2
This week we have a reading group meeting, attendees should have read the chapter and come prepared to discuss.
NYU students have free electronic access to the book here.
Note the special date and time: Wednesday December 2nd at 2:15pm.
February 2nd, 2010 - Michael Collins, MIT
TAG-based Structured Prediction Models for Parsing and Machine Translation
Abstract
February 9th, 2010 - Afshin Rostamizadeh, Courant Institute
Automatic Kernel Selection: Theory and Algorithms
Abstract
February 16, 2010 - Holiday, no seminar
February 23rd, 2010 - Afshin Rostamizadeh, Courant Institute
Automatic Kernel Selection: Theory and Algorithms, Pt. 2
Abstract
March 2nd, 2010 - Li Wan, Courant Institute
Maximum Margin Clustering
Slides -- Paper
March 9th, 2010 - Ameet Talwalkar, Courant Institute
Matrix Approximation for Large-Scale Learning
Abstract
March 23rd, 2010 - Stephen Boyd, Stanford University
Recent Advances in Convex Optimization
Abstract
NOTE: This talk will be at Google at 11:00am. Please contact by the 22nd Afshin to register as a visitor.
March 30th, 2010 - No seminar this week.
April 6th, 2010 - Mark Tygert, Courant Institute
Obvious yet overlooked means for testing statistical theories: an informal presentation
Abstract
April 20th, 2010 - No seminar this week.
Special Date: May 3rd, 2010 - Ameet Talwalkar, Courant Institute
PhD Defense: Matrix Approximation for Large-scale Learning
Abstract

2008-2009 Schedule

September 16th, 2008 - Afshin Rostamizadeh, NYU
Sample Selection Bias Correction Theory
Abstract Slides
September 23rd, 2008 - Sanjiv Kumar, Google
Sampling-based Approximate SVD
Abstract
September 30th, 2008 - No seminar
October 7th, 2008 - Ameet Talwalkar, NYU
Discussion: On the Margin Explanation of Boosting Algorithms, Wang et al. (COLT 2008)
Slides
October 14th, 2008 - Academic holiday, no seminar
October 21st, 2008 - Eugene Weinstein, NYU
Introduction to Topic Models
Abstract Slides
October 28th, 2008 - No seminar
November 7th, 2008, 11:30am, WWH 1302 (special session: please note unusual time and location) - Yishay Mansour, Tel-Aviv University and Google
Domain Adaptation with Multiple Sources
Slides
November 11th, 2008 - Ashish Rastogi, Google
Discussion: Does Unlabeled Data Provably Help? Worst-case Analysis of the Sample Complexity of Semi-Supervised Learning, Ben-David et al. (COLT 2008)
Slides
November 18th, 2008 - Afshin Rostamizadeh, NYU
Rademacher Bounds for Beta-mixing Distributions
Abstract Slides
November 25th, 2008 - No seminar
December 2nd, 2008 - Rob Fergus, NYU
Spectral Hashing
Abstract
January 20th, 2009 - Umar Syed, Princeton University
Hybrid Supervised/Reinforcement Learning Problems
Abstract Slides
January 27th, 2009 - No seminar
February 3rd, 2009 - Special joint session with the Cryptography Reading Group
Claire Monteleoni, Columbia University
Advances in Privacy-Preserving Machine Learning
Abstract Slides
February 10th, 2009 - Ameet Talwalkar, NYU
Discussion: Spectral Clustering with Perturbed Data, Huang et al. (NIPS 2008)
February 17th, 2009 - John Langford, Yahoo
PAC-Bayes
Tutorial Link Slides
February 24th, 2009 - Seminar canceled
March 3rd, 2009 - Eugene Weinstein, NYU
Discussion: Support Vector Method for Novelty Detection (Scholkopf et al., NIPS 2000)
Slides
March 10th, 2009 - No seminar
March 17th, 2009 - Spring break, no seminar
March 24th, 2009 - Afshin Rostamizadeh, NYU
Overview on Canonical Correlation Analysis
Paper Link
March 31st, 2009 - Ameet Talwalkar, NYU
Logistic Regression, Kernel Logistic Regression and Import Vector Machines
Paper Link
April 7th, 2009 - Aryeh Kontorovich, Weizmann Institute
Universal Kernel-Based Learning with Applications to Regular Languages
Abstract
April 14th, 2009 - No seminar
April 21st, 2009 - Spencer Greenberg, NYU
Occam' s Razor : Understanding the Cost of Complexity
Abstract

2007-2008 Schedule

September 4th, 2007 - Mehryar Mohri
Introduction to Rademacher Complexity (Part I)
Slides
September 11th, 2007 - Mehryar Mohri
Introduction to Rademacher Complexity (Part II)
Slides
September 18th, 2007 - No seminar
September 25th, 2007 - Ashish Rastogi
Title: McDiarmid's Inequality and its Applications
Abstract Slides
October 2nd, 2007 - Eugene Weinstein
Discussion: Search-based Structured Prediction, by Hal Daume, John Langford, and Daniel Marcu (submitted to Machine Learning)
Slides
October 9th, 2007 - Afshin Rostamizadeh
Discussion: Learning the Kernel Matrix with Semidefinite Programming, Lanckriet et al. (JMLR 2004)
Slides
October 16th, 2007 - Cyril Allauzen
Discussion: On Optimal Learning Algorithms for Multiplicity Automata, L. Bisht, N. Bshouty and H. Mazzawi (COLT 2006)
Slides
October 23rd, 2007 - No seminar
October 30th, 2007 - Ameet Talwalkar
Discussion: On the Nystrom Method for Approximating a Gram Matrix for Improved Kernel-Based Learning, Petros Drineas and Michael Mahoney (JMLR 2005).
Slides
November 6th, 2007 - Yishay Mansour
Reinforcement Learning, Part 1
Abstract   Slides   Class Notes
November 13th, 2007 - Yishay Mansour
Reinforcement Learning, Part 2
Abstract
November 20th, 2007 - No seminar
November 27th, 2007 - Afshin Rostamizadeh
Stability Bounds for Non-i.i.d. Processes
Paper Link
December 4th, 2007 - No seminar
December 11th, 2007 (NOTE ROOM CHANGE: WWH 1013) - Boulos Harb
Sampling Algorithms for Lp Regression and an Application to Feature Selection
Paper Links: Sampling Algorithms and Coresets for Lp Regression (SODA 2008), Feature Selection Methods for Text Classification (KDD 2007)
January 22nd, 2008 - Ameet Talwalkar
Title: Large-Scale Manifold Learning
Abstract
January 29th, 2008 - No seminar
Feburary 5th, 2008 - Afshin Rostamizadeh
Discussion: Learning Bounds for Domain Adaptation, Blitzer et al. (NIPS 2007)
Slides
February 12th, 2008 - Subhash Khot
Title: Hardness Results for Some Learning Problems
Abstract
February 19th, 2008 - Eugene Weinstein
Discussion: Discriminative Log-Linear Grammars with Latent Variables, Petrov and Klein (NIPS 2007)
Slides
February 26th, 2008 - Ashish Rastogi
Discussion: FilterBoost: Regression and Classification on Large Datasets, Bradley and Schapire (NIPS 2007)
Slides
March 4th, 2008 - no seminar
March 11th, 2008 - Rob Fergus
Title: Large image databases and small codes for object recognition
Abstract
Slides
March 18th, 2008 - Spring Break, no seminar
March 25th, 2008 - Sanjoy Dasgupta
Title: Random projection trees and low dimensional manifolds
Abstract
April 1st, 2008 - Afshin Rostamizadeh
Discussion: A Conditional Random Field for Discriminatively-trained Finite-state String Edit Distance and An Introduction to Conditional Random Fields for Relational Learning
April 8th, 2008 - Ameet Talwalkar
Discussion: Dimension Reduction Using Stable Random Projections
April 15th, 2008 - Risi Kondor
The skew spectrum of graphs --- a new class of graph invariants
Abstract
Slides
April 22nd, 2008 - Eugene Weinstein
Discussion: Discriminative Training of Decoding Graphs for Large Vocabulary Continuous Speech Recognition (Kuo, Kingsbury, Zweig, ICASSP 2007)
Slides
April 30th, 2008 - Ashish Rastogi
Special session: thesis defense

Details